Skip to content

Pull requests: NVlabs/cuda-oxide

New pull request New
16 Open 75 Closed
16 Open 75 Closed
Author
Filter by author
Loading
Label
Filter by label
Loading
Use alt + click/return to exclude labels
or ⇧ + click/return for logical OR
Projects
Filter by project
Loading
Milestones
Filter by milestone
Loading
Reviews
Filter by reviews
No reviews Review required Approved review Changes requested
Assignee
Filter by who’s assigned
Assigned to nobody Loading
Sort
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated Best match
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Pull requests list

feat: Add cbrt intrinsic support for f32 and f64 codegen Device code-generation pipeline (Rust MIR to IR to PTX) enhancement New feature or request intrinsics Device intrinsics and libdevice math mappings
#148 opened Jun 9, 2026 by goog00 Contributor Loading…
2 tasks
fix(mir-lower): declare direct __nv_* libdevice externs at the call site bug Something isn't working intrinsics Device intrinsics and libdevice math mappings IR-lowering Lowering between dialects (dialect-mir to LLVM dialect)
#144 opened Jun 8, 2026 by haixuanTao Loading…
fix(mir-importer): preserve op location/span in lowering error messages codegen Device code-generation pipeline (Rust MIR to IR to PTX) QoL-improvement Quality-of-life improvement; not a correctness fix
#143 opened Jun 8, 2026 by haixuanTao Loading…
fix(mir-importer): lower libm float ops to libdevice intrinsics bug Something isn't working codegen Device code-generation pipeline (Rust MIR to IR to PTX) intrinsics Device intrinsics and libdevice math mappings
#142 opened Jun 8, 2026 by haixuanTao Loading…
fix(mir-lower): fall back to live operand type in arithmetic signedness check bug Something isn't working codegen Device code-generation pipeline (Rust MIR to IR to PTX) IR-lowering Lowering between dialects (dialect-mir to LLVM dialect)
#141 opened Jun 8, 2026 by haixuanTao Loading…
fix(mir-importer): translate str, fn-pointer and fn-item types bug Something isn't working codegen Device code-generation pipeline (Rust MIR to IR to PTX)
#140 opened Jun 8, 2026 by haixuanTao Loading…
Add scalar f16 atomic support cuda-feature A CUDA hardware/toolkit feature (TMA, tcgen05, WGMMA) enhancement New feature or request intrinsics Device intrinsics and libdevice math mappings
#134 opened Jun 7, 2026 by advpropsys Contributor Loading…
1
feat(codegen): fma contraction and an opt -O3 pass to match nvcc defaults codegen Device code-generation pipeline (Rust MIR to IR to PTX) enhancement New feature or request perf Performance of generated code or of the compiler itself
#117 opened Jun 6, 2026 by alejandro-soto-franco Contributor Loading…
Add debug information to the generated PTX code debug info Debug information in generated PTX/IR (line/source maps) enhancement New feature or request
#106 opened Jun 3, 2026 by Barrow099 Contributor • Draft
Add support for cargo oxide debug and pipeline commands in non-example projects cargo-oxide The `cargo oxide` subcommand and build automation enhancement New feature or request
#103 opened Jun 2, 2026 by Barrow099 Contributor Loading…
Support typed NVVM IR for pre-Blackwell libNVVM targets Depends Blocked on another PR/issue or an upstream dependency enhancement New feature or request llvm-export Textual LLVM/NVVM IR exporter (llvm-export crate)
#101 opened May 30, 2026 by mohamedsamirx Contributor Loading…
1
1
Keep in-flight async future results alive on drop bug Something isn't working host-apis Host-side runtime APIs (cuda-host, cuda-core, cuda-async)
#100 opened May 29, 2026 by fallintoplace Loading…
sys: honor CUDA_TOOLKIT_PATH in runtime discovery bug Something isn't working build-related Build scripts, toolchain discovery, or compilation setup
#95 opened May 27, 2026 by yagna-1 Loading…
1
Fix/GPU memory leak on DeviceBuffer allocation failure bug Something isn't working host-apis Host-side runtime APIs (cuda-host, cuda-core, cuda-async)
#92 opened May 26, 2026 by winshaurya Contributor Loading…
cuda-bindings: handle CUDA header discovery errors bug Something isn't working build-related Build scripts, toolchain discovery, or compilation setup cuda-bindings Raw FFI bindings to the CUDA driver/runtime
#88 opened May 26, 2026 by yagna-1 Loading…
1
3
Add cvt_f16x2_f32 intrinsic for f32-to-f16x2 packing enhancement New feature or request intrinsics Device intrinsics and libdevice math mappings
#66 opened May 16, 2026 by honeyspoon Loading…
ProTip! Add no:assignee to see everything that’s not assigned.